Download Tilburg University Theories on the scope for foreign exchange

Tilburg University
Theories on the scope for foreign exchange market intervention
Almekinders, G.J.
Publication date:
1993
Link to publication
Citation for published version (APA):
Almekinders, G. J. (1993). Theories on the scope for foreign exchange market intervention. (CentER Discussion
Paper; Vol. 1993-42). CentER.
General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners
and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.
- Users may download and print one copy of any publication from the public portal for the purpose of private study or research
- You may not further distribute the material or use it for any profit-making activity or commercial gain
- You may freely distribute the URL identifying the publication in the public portal
Take down policy
If you believe that this document breaches copyright, please contact us providing details, and we will remove access to the work immediately
and investigate your claim.
Download date: 09. mei. 2017
~~
CBM
R
8414
1993
42
1 scussion
~
~
for
~~c Research
a er
IIINININII ~III~ InN hnl I~n q101~ Ilu
CentER
for
Economic Research
No. 9342
Theories on the Scope for Foreign
Exchange Market Intervention
by Geert J. Almekinders
June 1993
ISSN 0924-7815
' ~- - -~ ~;~.~
-
~
i ~ ,~~,~!' ~ . ~~ . F3
' ~~~'~'
~ ~~
~ ~'
~~3i.éC~THEEK
TÍLBIUF~G
THEORIES ON THE SCOPE FOR FOREIGN EXCHANGE MARKET INTERVENTION
GEERT J. ALMEKINDERS'
Tilburg University
April 1993
Abstract
This
paper
traces
out
recent
developments
in
modeling
foreign
exchange
market
intervention. The central question is whether and how intervention is able to influence the course
of the exchange rate. The channels of inFluence of unsterilized and sterilized intervention in
some well-established
models of exchange
rate determination
are set out.
Furthermore,
the
mechanics of foreign exchange market intervention in some recent studies are analyzed. These
studies adhere to the assumption that the foreign exchange market is efficient in the sense that
market participants use some structural
model as a yardstick when taking positions on the
foreign exchange market. Finally, some altemative approaches to the study of foreign exchange
intervention are discussed. The latter approaches drop the assumption that the foreign exchange
market is efficient.
'I wish to thank J. Sijben and S. Eijffinger for helpful comments on an earlier draft of this
paper. My appreciation does not implicate them for errors either of omission or commission that
remain in this paper.
I INTRODUCTION
During the period 1985-1990 the world's foreign exchange mazkets experienced recurrent
strains. The central banks of the main industrialized countries reacted by intervening in the
foreign exchange markets on a scale not seen since the demise of the Bretton Woods system of
fixed exchange rates in 1973. This has led to a renewed interest in studying the scope for central
bank foreign exchange market intervention. Recent empirical investigations into the objectives
and effectiveness of foreign currency operations by central banks are surveyed in Almekinders
and Eijffinger (1991) and Edison (1992). This article traces out recent developments in modeling
foreign exchange market interventions. lt does not address the question of whether central bank
intervention is desirable (cf. Pilbeam
1991, ch.l). Instead it starts from the assumption
that
central bank intervention is a given feature of foreign exchange markets.' The central question
is whether and how intervention is able to influence the course of the exchange rate. As a consequence, no attention is paid to exchange rate management as an instrument of (domestic income)
stabilization policy (Marston I985, Pilbeam 1991, ch.2 and Alogoskoufis 1993).
The remainder of this article is organized as follows. Section II provides a definition of
exchange market intervention and introduces the concept of sterilization. Section III and IV set
out the channels of influence of unsterilized and sterilized intervention in the well-established
models of exchange rate determination depicted in the left-hand column of Table l.Z Furthermore, the mechanics of foreign exchange market intervention in the studies shown in the two
right-hand columns of Table 1 are analyzed. Section V discusses some altemative approaches to
the study of foreign exchange intervention. Section VI concludes.
II INTERVENTION AND STERILIZATION: THE BASICS
I define an exchange market intervention as a sale or a purchase of foreign currency by
the domestic monetary authorities aimed at changing the exchange rate of the domestic currency
vis-à-vis one or more foreign currencies.
An important
distinction
is to be made behveen
unsterilized or 'monetary' interventions and sterilized or 'pure' interventions. The efiect of both
rypes of intervention will be illustrated by pointing to the changes in the balance sheet of the
central bank and the private sector each brings about.
A srylized balance sheet of the domestic central bank reflects that the monetary base of
the domestic country (.A~fB), which consists of total domestic currency
in circulation and the
reserves of the private banking system, is equal to the sum of net foreign assets (,vFA)
and
domestic assets (DA) in the hands of the domestic central bank.
'The implementation of money market and foreign exchange market policy in the Netherlands is aptly described in Den Dunnen and De Wilde (1992).
'For detailed surveys of exchange rate theories see Visser ( 1989) and MacDonald and Taylor
(1992).
2
Table 1 Representative Exchange Rate Models and Some Selected "Intervention" Studies
Exchan e rate model
Purchasin
Non-sterilized intervention
Sterilized intervention
Black (1985)
Black (1985)
D'a'ic 8r Bazzoni (1992)
Natividad á Stone (1990)
Moreno 8r Yin (1992)
Masson 8c Blundell-Wi
Power Pari
Mundell-Flemin
Model
Flex- rice Monet
Model
Stick - rice Monet
Model
Portfolio Balance Model
all (1985)
Stock-Flow Portfolio Model
Balance sheet oJthe domestic centra! bank
Assets
Net foreign Assets (A'FA)
Gold
Foreign currency
SDR
Net position in IMF
Liabilities
Mooetary Base (MB)
Total currency in circulation
Reserves of the private banking system
Domestic Assets (DA)
Govemment securities
Loans to commercial banks
A purchase (sale) of foreign currency by the domestic central bank from the domestic
private sector leads to a direct increase ( decrease) in the domestic money supply. In terms of
changes in the above balance sheet's entries: ,~t~ - p,yFq .'
Nowadays, the main industrialized countries implement monetary polic} based on the
control of some monetary aggregate. The monetary authorities of these countries do not want
exchange rate policy to interfere with monetary policy. Thus, the domestic central bank may
'The exact implications of an operation in foreign currency by the domestic central bank
depend on who is the counterparry in the transaction. An earlier version of this paper which was
circulated as a Research Memorandum of the Department of Economics, Tilburg University and
which is available from the author on request distinguishes between the domestic commercial
banking system and the domestic non-bank private sector. Obviously, in theoretical models
without a fractional reserve private banking system this distinction is not relevant. The earlier
version of this paper also elaborates on the intemational implications of an official intervention.
According to most standard expositions on foreign exchange intenention official transactions
only affect the balance sheets of the domestic central bank and those of the domestic private
sectoc As clearly put forward by Weber (1986) and Belongia (1992), intervention conducted by
the domestic central bank almost necessarily involves the actions of the foreign central bank.
attempt to adhere to a preannounced target for the growth rate of some monetary aggregate. To
do so, it can sterilize the effect of the initial exchange market intervention by selling long-term
domestic assets denominated in domestic currency, say govemment bonds, from its openmarket
portfolio to the domestic private sector leaving the domestic monetary base, ceteris paribus,
unchanged in spite of the initial purchase of foreign currency. The effect of the exchange market
intervention on the monetary base is completely neutralized when
balance sheet of the domestic central bank.
ppA --ONFA
in the
Obviously, in the end, there is no difference between domestic monetary policy on the
one hand and unsterilized intervention on the other hand.' For that reason, it is often argued that
unsterilized
intervention amounts to using the foreign exchange market to conduct monetary
policy in lieu of the domestic financial market. I opt for a more positive approach of unsterilized
intervention. While in general the effect of central bank intervention on the reserve position of
domestic commercial banks is sterilized routinely, more emphasis on limiting exchange rate
volatility in the formulation of monetary policy will most likely be reflected in a lower degree of
sterilization. Therefore, 1 find it useful to review the possibilities and limitations of uns[erilized
intervention working through what Humpage (1986) calls the monetary channel.
When the monetary authorities take action to neutralize the money market effect of the
initial purchase of foreign currency this monetary channel is ruled out. Changes in the domestic
central bank's balance sheet imply that sterilized or 'pure' intervention involves an exchange of
foreign
currency
denominated
securities
in
return
for
govemment
bonds
denominated
in
domestic currency, however. In a closed financial system, the domestic private sector experiences the reverse swap of assets in its portfolio. Thus, the currency composition of the domestic
private sector's portfolio of investments changes. Below it will be analyzed whether and how
this forced change in the curtency composition of private investors' portfolios has repercussions
on the course of the exchange rate through a portjolio channel.
III THE FLOW' APPROACH TO E~CCHANGE RATE DETERtiII,~iATION
III.1 Purchasing power parity
Since long it has been recognized that purchasing power parity (PPP) and hence the
relative price of national outputs provides the international financial system with a long term
anchor for exchange rate movements. The absolute version of purchasing power pariry relates
the exchange rate of the foreign currency in terms of domestic currency to overall price levels in
'Domestic monetar} policy typically involves the same change in the domestic monetary
base brought about by the central bank via an open market purchase or sale of domestic govemment securities (l1,tifB - (1DA ).
4
Purchasing power paziry as a theory of exchange rate determination
(11
S, - P, I P,'
absolute purchasing power pariry
(~)
S, - P - P,~
relative purchasing power parity
(3)
M,~'~ V,~'~ - P,''~ Y,~'~
quantity theory of money
(4)
S, - (M - M ~ ), t (V - ~' ), - (Y - Y ~ ),
solution for the exchange rate
where S. P. M. tr and
of foreign currency), the
money and real income,
the variable refers to the
percentage change.
y are the exchange rate ( measured as domestic currency units per unit
price level, the quantity of money supplied, the income velocity of
respectively. Subscript t denotes time. An asterisk (- ) indicates that
same variable in the foreign countr}. A dot over a variable denotes a
the domestic and foreign country. It is an extension of the well-known Law of One Price to
general price levels. Accordingly, goods arbitrage equalizes the market exchange rate to the
purchasing power parity rate given by equation (l). Relative purchasing power pariry is simply
the concept of absolute PPP expressed
in growlh rates.
It asserts that the value of foreign
currency in tetms of domestic currency tends to rise at a rate equal to the difference between
domestic and foreign inflation (equation (2)).'
A further expression for the percentage change in the domestic currency value of foreign
exchange can be obtained by combining the PPP approach with the Quantity Theory of Money
for both the home and the foreign country (equation (3)). With the income velocity of money
and the level of real income unchanged,
an increase
in the domestic money supply will be
reflected in higher domestic prices. Hence, the depreciation of the domestic currency (rise in S,)
implied by equation (4). As a direct consequence it follows that an unsterilized intervention will
in the long run, ceteris paribus, influence the level of the exchange rate. By contrast, a pure
intervention
wfiich,
by definition,
lacks
a
money
market
effect
leaves the
exchange
rate
unaffected.
III.2 The Mundell-Fleming model
IIL2.1 The srandard dlundell-Fleming model and inrervention
'In generai, the short-run validity of purchasing power parity is rejected in empirical tests
(for a recent survey, see Giovannetti (1992)). This is because in practice short-tertn capital flows
swamp the trade balance of the balance of payments. Thus, short-term exchange rates are set in
financial markets rather than in goods markets. Currency prices fluctuate minute by minute.
Price levels, in contrast, are sticky and adjust slowly. With exchange rates moving more quickly
than goods prices, deviatíons from PPP will arise. ï'herefore, exchange rates can persistently
deviate from their purchasing power parity-implied values.
5
Particularly since ihe collapse of the Bretton Woods system of fixed but adjustable
exchange rates a lot of efforts have been devoted to investigate the determinants of short-term
fluctuations in the exchange value of foreign currencies. Initially, most of these investigations
were carried out along the lines of the Mundell-Fleming model. Mundell (1963) and Fleming
(1962) are seminal articles exploring the possibilities and effects of monetary and fiscal policy
under fixed and flexible exchange rates. They were written at a time when the Bretton Woods
fixed
exchange
rate
system
was still
operative.
The
theoretical
framework
underlying
the
analysis in the two studies has become known as the Mundell-Fleming model. Although it is not
the primary aim of this model to explain exchange rate movements, it represents an authoritative
formulation of the flow approach to exchange rate determination.`
Central to the Mundell-Fleming model when looked at as a theory of exchange rate
detertnination is the balance of payments equilibrium condition. It states that with frrely floating
exchange rates the net volume of foreign currency flowing through the currenct account and the
capital account sums up to zero. The basic model assumes static expectations. With the expected
exchange rate depreciation equal to zero, a retum differential between domestic and foreign
bonds and hence net intemational capital flows can only arise through a difference between the
domestic rate of interest and the given world interest rate.
Imagine the domestic economy is initially in equilibrium. In section fI it was established
that an unsterilized purchase of foreign currency form the domestic private sector conducted by
the domestic central bank increases the domestic money supply. Although the domestic interest
rate
is de jacto fixed at the foreign level due to the assumed perfect
international
capital
mobility, this can be thought of as causing an incipient decline in the domestic interest rate
which in turn leads to a capital outflow and a lower value of the domestic currencv.
In section
II
it was established
that a pure
intervention
results in a change in the
currency composition of the private sector's nominal financial wealth but not in its volume. As
bonds denominated in different currencies are perfect substitutes in the Mundell-Fleming model
the altered currency composition does not have any effect on the exchange rate.
II12.2
(Un)sterilized intervention in Black's (1985J flox~ mode!
Not every author surrenders to the apparent inefiectiveness of sterilized intervention in
flow models of the exchange rate. With the benefit of hindsight Black ( 1985) combines some
selected building blocks of the various models of exchange rate determination which survived in
empirical testing. Black argues that, in general, empirical evidence is not supportive of short-run
PPP (see also footnote 6) and perfect substitutability of assets denominated in different currencies. He assumes imperfect substitutability and rational expectations under imperfect information.
The author starts out by deriving a short-run stock equilibrium condition for foreign assets.
óFor a solid representation of the Mundell-Fleming model, see MacDonald (1988, Ch.3).
6
Within a mean-variance framework he arrives at an expression for the net stock of bonds
denominated in foreign currency f private investors are willing to hold:
i, - a[(Err,., - s,) t(i,'
(~)
- i,)]
wrlh ~3 -
I
P Q,. ~
Accordingly, the willingness of speculators to hold foreign assets depends positively on the
expected depreciation of domestic currency (E ~ i- s) and the excess of the foreign over the
domestic interest rate (i '
and the
investors'
- i). It is inversely related to the variability of the exchange ratea~ ~
degree of risk
aversion
p.
Obviously,
the Mundell-Fleming
model
is
concemed with the equilibration of flow-demand and supply for foreign exchange. The first
difference of the stock of foreign assets held by speculators in consecutive periods is taken to
imply the outflow of private capital
measured
in foreign currency.
~f
(equation
( 1)).
Equation ( 3) accounts
Equation ( 2)
for changes
is the trade balance
in central bank reserves.
Accordingly, the central bank intervention reaction function consists of a two parts. Firstly, the
"leaning against the wind" component (- ~
(s
- s i)) reflects that the central bank buys foreign
cutrency ( positive value of ~) when the current exchange rate is lower than the exchange rate
in the previous period.
Secondly, a component
capturing
exchange rate closer to the target level s r (-i;2 (s
intervention aimed at driving the
- s~) reflects that the central bank buys
foreign currency ( positive value of ~r ) when the current exchange rate is lower than this target
level. Equation ( 4) is the balance of payments equation.
Nonsterilized purchases of foreign currency increase the domestic monetary base and
lower the domestic interest rate. In Black's model this leads to additional private purchases of
foreign bonds and, hence, to a strengthening of the initial effect of the central bank's purchases
on the balance of supply and demand in the foreign exchange market. However, in Black's view
the monetary channel of intervention is not the most interesting one.
The analysis of the effectiveness of sterilized intervention is far from casual. Black adThe Mundell-Fleming model amended as in Black (1985)
OJ - Q ~(Err,., - s,) -(E,-~s, t s,-~) . 4i,' - Or,] ~ Av,
t, - T(s, ' P,~ - P,) ' E,
net flow of private capital
(2)
(3)
0 r, ' -5i (s, - s,-i) - ~, (s, - s ~
intervention reaction funct.
(4)
Or,.l, t,~f -0
balance of payments
(I)
trade balance
where f and r are the stock of net foreign assets held by the private sector and the central
bank, respectivély. w and E are uncorrelated random disturbance terms with mean zero,
constant variance and zéro autocorrelation; t, ~ and ~ are positive constants; lowercase letters
refer to natural logarithms of variables; 0 is t é first-difference operator with p x, - x, - x,.~:
E represents the expectations operator conditional on information available at time l;
heres to the (perhaps heroic) assumption that private speculators have a stabilizing
inFluence on
the course of the exchange rate. Therefore, for the purpose of his study it is not the direct
influence of intervention on [he exchange rate that counts. The crucial issue is whether
the
monetary authorities are able to induce private investors to assume a larger position in foreign
currency. Thus, Black is in search of an indirect effect of sterilized intervention. As investors are
assumed to be rational they know the equilibrium value of the exchange rate. Thus sterilized
interventions are effective if they reduce uncertainty among investors (measured by 6~ ~) so as to
mobilize stabilizing speculative capital flows.
Based on the model described by the equations (I)-(4), which is quite similar to the one
analyzed in Neumann (1984), Black derives certain expressions for the willingness of speculators
to bear risk R. It should be noted that the author assumes the foreign interest rate to be constant
and, more importantly, that sterilized interventions leave the domestic interest rate unaffected.
He concludes that both "leaning against the wind" intervention (~
, p) and target intervention
(~2 , p), the latter provided that the target chosen by the monetary authorities is equal to the
purchasing power parity-implied long-run equilibrium value, are effective according to his own
definition. Both t}~pes of sterilized intervention appear to reduce the level of uncertainty about
the exchange rate Q?~ and hence increase the volume of speculators' stabilizing position taking
in foreign currency. The underlying mechanism by which this is accomplished remains rather
nebulous. Of course, the central bank's transactions in foreign exchange initially have a bearing
on the flow equilibrium. The stabilizing impact of the official operations in foreign currency
might remove part of the uncertainty among private speculators and hence strengthen the initial
effect of the official transaction. However, the central bank is only a minor participant on the
market for foreign exchange. A huge amount of intervention may be required to convince private
market participants. Then, sterilization involves voluminous offsetting open market transactions.
For instance, when the central bank initially buys foreign exchange from the domestíc non-bank
private sector to bring the exchange rate closer to the PPP level from below, sales of domestic
currency denominated government bonds are required to neutralize the money market effect of
the intervention. The price of domestic bonds will decline and the interest rate will go up.
Hence, the assumption of constant domestic and foreign interest rates, which is at the heart of
the analysis of unsterilized intervention in Black (1985), seems to be questionable.
IV. THE ASSET MARKET APPROACH TO EXCHANGE RATE DETERMINATION
IV.1 Introduction
The flow analysis according to Mundell-Fleming views the exchange rate as the relative
price of natíonal outputs. The enormous growth of financial markets tumovers in general and the
surge in short-term international capital Flows in particular cause a country's current account of
8
the balance of payments to be swamped by the capital account. While short-term exchange rates
are set in financial markets rather than in goods markets, the exchange rate is more and more
viewed as an asset price which is equal to the price of one national money in terms of another.
Furthermore, according to this approach exchange rate changes are not caused by shifts in the
typical demand and supply schedules for foreign exchange because of real transactions; changes
in the perception of the market as a whole with regard to the value of one currency vis-à-vis one
or more other currencies are the crucial factor.
IV.2
The Oexible-price monetary model
One branch of asset market models of exchange rate determination assumes that wealth
holders are indifferent as [o the proportions of domestic and foreign currency denominated assets
in their portfolios given that they yield the same retum expressed in one currency.
In other
words, portfolio shares are infinitely sensitive to changes in expected rates of return. Hence, the
perfect substitutabiliry hypothesis implies that otherwise identical bonds denominated in different
currencies can be viewed as one homogeneous asset. It follows that under risk-neutrality uncovered interest parity prevails (equation (3)) and the world bond market always clears instantaneously. Furthermore, purchasing power pazity is assumed to hold continuously (equation (4)). This
implies that the real exchange rate and, thus, the relative price of domestic and foreign goods is
constant over time due to perfect international arbitrage on the goods market. Consequently, in
the flexible-price monetary exchange rate model demand and supply conditions on the markets
for goods and bonds are irrelevant; in the short run a bilateral exchange rate is detertnined by the
requirement of money market equilibrium in the two countries involved whereby it is assumed
that residents of each country hold only their own money (equations (1) and (2)).
Under the assumption of flexible prices, the equations ( I}(4) can be solved for the ex-
The flexible-price monetary model
home money market equilibrium
(I )
m- p t~y - li
(2)
m,'
(3)
i, - i,' t E~[s,-~] - s,
uncovered interest parity
(4)
s, - P, - P,'
purchasing power parity
(5)
sI
~[ ~]~ E[(m
~
Í 4Íl ~.0 Í t?.
- p," f~'y,'
foreign money market equilibrium
-~.'i,'
- m,: )-~(y,,:, - y ) ]
solution for the exchange rate'
where ~ and ~ are the elasticity and the semi-e(asticity of money demand with respect to
income and the interest rate, respectively. Here these are assumed identical for both countries.
'For a lucid exposition on the derivation of this solution for the exchange
monetary model with rational expectations, see Visser (1989, p. 18-22).
rate
in the
9
change rate to arrive at equation (5). [t states that the bilateral nominal exchange rate depends on
the current and expected future values of relative money supplies and relative outputs in both
countries. Thus, the flex-price monetary approach to the exchange rate predicts that an unsterilized purchase of foreign bonds from the domestic non-bank private sector conducted by the
domestic central bank leads to a rise in the domestic currency price of foreign exchange. The
purchase of foreign bonds causes the domestic money supply to increase. Through equation (1),
the inherent excess supply of money is wiped out by an instantaneous rise in the domestic price
level. Equation (4) implies that the domestic currency price of foreign exchange goes up as a
consequence of the initial unsterilized purchase of foreign bonds.
Furtherrnore and perhaps more surprisingly, based on the expression in (5) some authors
argue that sterilized intervention can be expected to alter the current exchange rate through what
is called the expectations channel oJintervention. Sterilized intervention can provide private exchange market participants with new information or a signal about the future course of monetary
policy. The "signalling h}pothesis" was first proposed by Mussa (1981). Accordingly, a sterilized
purchase of domestic cun-ency from the domestic private sector signals an expansionary future
domestic monetary policy. The expectations of future looser domestic monetary policy will make
the domestic currency depreciate and hence the exchange rate go up, even though the initial
intervention's money market effect is neutralized in the short-run (see equation (5)).
The relevance of the signalling or expectations channel of sterilized intervention is not
undisputed. Private exchange market participants will only pay attention to the signal embodied
in the sterilized intervention when a stable relationship has emerged with interventions leading
changes
in monetary
policy
aimed
at
some
exchange
rate
objective.
Whether
this stable
relationship exists and whether private exchange market participants pay attention to it is an
empirical issue.e More importantly, for the signalling hypothesis to be valid central banks have
to back up interventions with subsequent changes in monetary policy. In other words, curtent
sterilized interventions pre-determine the path of future money grow2h and hence interfere with
monetary policy. The neutralization of the initial intervention's money market effect is limited to
the short run. This no Ionger meets the definition of sterilized intervention given above exactly.
[V.3 The sticky-price mooetary model
IV.3.l The standard sticlry-price monetary mode! and intervention
BKlein and Rosengren (1991) use newspaper reports on dollar intervention by the United
States and Germany. They find that interventions did not precede changes in monetary policy
and periods of active intervention were not followed by monetary policy changes. Kaminsky and
Lewis (1992) strongly reject the hypothesis that interventions convey no signal. However, they
also find that in some episodes, intervention signalled changes in monetary policy in the opposite
direction of the conventional signalling story.
10
The assumptions underlying the flexible-price monetary model are not compatible with
the persistent rejection of purchasing power parity in empirical tests (see footnote 6). Dombusch
(1976) amends the flex-price monetary model to take account of this empirical regularity. In the
resulting sticky-price monetary model goods prices initially do not respond to disturbances of the
goods market equilibrium.9 With the abandonment of short-run purchasing power parity, an
equation to explain the evolution of the price level is required. Dombusch assumes that the price
level adjusts in proportion to excess demand. This process continues until (long-run) purchasing
power parity is restored.
The classical exercise in the Dombusch-model analyzes the effect of a monetary expansion in the domestic country. This happens to be compatible with an unsterilized purchase of
foreign currency by the domestic central bank. In the short run, with the domestic prices still
unchanged, the increase in real money balances induces a decline in the domestic interest rate.
This is the immediate liquidiry effect of the initial unsterilized purchase of foreign currency. The
money market effect of this intervention makes investors expect a long-run depreciation of the
domestic currency. The decline in the domestic interest rate plus the expected depreciation of
domestic currency seriousl} detract from the relative attractiveness of domestic bonds. Speculators want to be compensated for borh factors. Consequently, the instantaneous restoration of
goods and money market equilibrium after the monetary shock requires the exchange rate to
overshoot
its long-run value. The initial real depreciation (the exchange rate rises while the
domestic goods price remains constant) and the lower domestic interest rate increase the demand
for domestic goods. This will cause domestic prices to rise and the economy gradually moves to
the new equilibrium. In sum, with the price level sticky, the system can not jump to the new
long-run equilibrium. Instead, the exchange rate jumps, placing the domestic economy onto the
stable path to the new long-run along which the domestic price level and the exchange rate
appreciates.
The effectiveness of foreign exchange market intervention in both the flexible-price and
the sticky-price monetary model depends crucially on its money market effect. Once this is
neutralized not a single economic variable in the model is affected by the intervention operation.
For instance, offsetting open market sales of domestic currency denominated government bonds
to sterilize the money market effect of an initial purchase of foreign currency lead to an incipient
rise
in the
domestic
interest
rate.
As otherwise
identical
bonds
denominated
in different
currencies are perfect substitutes, the demand for domestic currency denominated govemment
'A theoretical explanation for this phenomenon can be found in Okun (1981). He distinguishes customer markets from auction markets. The market for domestic goods is a customer
market on which the price is the result of an ongoing relationship between buyer and seller.
Therefore it is costly to change this price. Money and bonds are traded in auction markets on
which buyers and sellers are more or less anonymous. Moreover, market participants are used to
prices changing in real time.
Il
bonds is perfectly elastic. The domestic interest rate does not have to rise for investors to be
willing to hold the additional supply of these bonds. Put differently, when the economy is
initially in a steady state the expected change in the exchange rate is zero. Consequently, the
current exchange rate is equal to its long-run equilibrium value. Due to the lack of a money
market effect, pure interventions do not alter the long-run equilibrium exchange rate. Hence pure
interventions preserve a zero expected exchange rate change. Then, uncovered interest pariry
implies that interest rates at home and abroad have to remain equal.
IV.3.2
Unsterilized intervention in Djajic and Bazzoni's (1992J sticky-price model
The
undisputed.
effectiveness
of monetary
intervention
in the sticky-price
monetary
model
is
Djajic and Baz~oni (1992) modify Dombusch' (1976) model to include a rule
governing unsterilized foreign exchange market intervention (equation (1)). This rule describes
the reaction pattem of the monetary authorities in response to shocks to the exchange rate.
Accordingly, the domestic money supply is brought down below its pre-disturbance level mo in
response to a depreciation of the domestic currency which has raised the exchange rate above its
pre-disturbace level so. The extent of the monetary contraction is reflected by the value of the
constant
w which is a policy-determined coefficient.
Obviously, the polar cases w- p and
w- oo are compatible with freely floating exchange rates and fixed exchange rates, respectively. Agents are assumed to be risk neutral, to know the structure of the model and to form their
expectations rationally (equation (3)). Accordingly, the actual rate of depreciation s equals the
expected rate of depreciation of domestic currency (s ~). Hence, the uncovered interest parity
condition can be written as in equation (4). With output fixed at the full emplo}~rtent level y, the
change in the price level is given by equation (5), where v is a shift parameter which reflects
commodity-market disturbances.
From the discussion of monetary
straightforward
intervention in the previous section,
it is rather
that the greater the magnitude of unsterilized "leaning against the wind"
intervention, the greater the extent to which pressures on the domestic currency to fall (rise) in
value, in response to, for example, an increase (decrease) in the foreign interest rate, are
absorbed through a reduction in (an expansion of) the domestic money supply rather than a
Unsterilized intervention in Djajic and Bazzoni's (1992) sticky-price model
0 s w ~ o0
intervention rule
( I)
m- mo - w (s - sa)
(2)
m - P ' ~Y - ~i
money market equilibrium
(3)
s - s `
rational expectations
(4)
i - i '
(5)
p - n (d -Y) - n ~S (s -P) i (Y -1) y - a i t v]
. s .
uncovered interest pariry
goods price dynamics
Iz
depreciation (an appreciation) of the domestic currency. Put differently, a move away from fixed
exchange rates (i.e., a lower value of w) allows for greater stability of monetary aggregates.
Djajic and Bauoni stress that these are long-run considerations. They analyze the
dynamic properties of the system in (1)-(5) for different values of w, the parameter capturing
the degree of "leaning against the wind" intervention. Rather than analyzing short-run exchange
rates over- or undershooting their long-run value, focus is on the evolution over time of the
money stock as a function of w, the degree of "leaning against the wind" intervention.
The crucial assumption in the Dornbusch model is that asset markets and exchange rates
adjust fast relative to the goods market and the price of domestic output. A dírect consequence
of this assumption is that, in the face of shocks impinging on the domestic economy, jumps
in
the exchange rate are required to achieve short-run equilibrium in the goods and assets markets.
[t is not very likely that these jumps in the exchange rate are compatible with the objectives
of
the monetary authorities. Hence, intervention is called for. Djajic and Baz~oni examine the
economy's adjustment to both a goods-market and an asset-market disturbance.
Initially, the economy is in a steady state. Consider the effect of an increase in the
demand for exports, reflected by an increase in v in equation (5). Firstly, under freely floating
exchange rates the increased demand for exports simply leads to an instantaneous appreciation of
the domestic currency which lowers exports to their original level. Secondly, in case the monetary authorities pursue a policy of 'leaning against the wind' they will partly resist the rise in the
value of the domestic currency. This necessitates a jump in the money supply to bring the
economy on the relevant stable path towards the new steady state. Along it, domestic prices rise
according to equation (5). The rising price level increases the demand for money. The concomitant rise in the domestic interest rate (equation (z)) results in renewed upwazd pressure on the
value of domestic currency (equations (4) and (5)). Due to the ongoing 'leaning against the
wind' policy this is partly translated into a gradually loosening monetary stance. Thirdly, under
fixed exchange rates the increase in the demand for exports will set the economy on a path of
gradually
rising prices. Along this path to the new steady state the money supply must be
increased in proportion to the increase in domestic prices to preserve the fixed exchange rate.
What stands out is the rather counter-intuitive finding of larger short-run movements.of
the nominal money supply under a'leaning against the wind' policy than under fixed exchange
rates. This result seems to be due mainly to the use of an asset market model of exchange rate
determination to analyze the effect of a real shock. This is fallacious while in asset market
models the exchange rate is essentially viewed as a monetary phenomenon. Djajic and Baz~oni
azgue that the increased demand for exports initially leaves the demand for money unaffected.
Then, within the Dombusch model, indeed there is no scope for an appreciation of the domestic
currency. I would argue that, to examine the implications of various exchange rate regimes for
the short-run effects of goods-market disturbances appropriately, flow models like the Mundell
Fleming model rather than asset market models offer the appropriate framework. In the former
13
models an increased demand for exports will cause an incipient appreciation of the domestic
currency. Then, under fixed exchange rates the monetary authorities are obliged to ease the
monetary stance immediately after the goods-market shock occurs.
The Dombusch model does offer an appropriate framework to analyze the implications
of various exchange rate regimes for the short-run effects of financial-market disturbances like
an increase in the foreign interest rate. The results presented by Djajic and Bazzoni for this casus
are much less powerful and in line with intuition. The positive shock to the foreign interest rate
induces a net outflow of capital. This leads to an (incipient) rise in the exchange rate. Official
purchases of domestic currency by the domestic central bank lead to an instantaneous contraction
of the money supply both under a'leaning against the wind' rule for intervention and under
fixed exchange rates.'o Thus, Djajic and Baz2oni arrive at the plausible result that the reduction
in the money supply in response to an increase in the foreign interest rate is larger under fixed
exchange rates than under a regime of managed floating.
IV.3.3 (Partly) sterilized inlen~enriorr in Natividad and Stone's (I990J sticky-price mode!
At the end of section IV.3.1 it was established that the sticky-price monetary model does
not provide any scope for the effectiveness of sterilized interventions. In spite of that it could be
of interest to investigate the implications of a varying degree of sterilization. Natividad and
Stone ( 1990) extend the original
Dombusch model to include separate policy functions
for
domestic credit and central bank foreign exchange reserves while allowing for variable sterilizatiou. The intervention reaction function ( equation ( I)) permits exogenous intervention ( where a
positive value of r
denotes a purchase of foreign currency by the domestic central bank) and
FY
endogenous responses to an observed gap between the contemporaneous and long-run equilibrium real exchange rate. The domestic credit reaction function ( equation ( 2)) allows for exogenous
operations ( where a positive value of cF-~, denotes monetary expansion through an open market
purchase of domestic currency denominated govemment bonds by the domestic central bank),
endogenous sterilization of a fraction x,
of the effect of changes in the central bank's foreign
exchange reserves and endogenous attempts to smooth deviations of the interest rate from its
long-run equilíbrium value. Equation ( 3) states that base money and, thus, the money supply is
equal to the sum of domestic assets ( c) and net foreign assets in the hands of the central bank
(r - nja~a). The interest pariry relationship is quite similar to the equation determining the
outflow of private capital
in Black ( 1985).
The net stock of bonds denominated in foreign
currency the domestic residents are willing to hold is given by
'oDjajic and Bazzoni show that for certain parameter values the jump in the money supply is
slightly larger in case of "leaning against the wind" intervention than in case of fixed exchange
rates. In my view this is a technical artefact of the model which defies economic explanation.
Furthertnore, it is not clear at all whether these parameter values imply a realistic degree of
"leaning against the wind" intervention.
la
(0)
J-nja-nja`9-all"-Íts`l
Thus equation (0) accounts for the fact that a country's net foreign assets can be in the hands of
the private sector and in the hands of the central banks (nfa`g - r). Rewriting (0) leads to
equation (6). W~en (3 goes to infinity, otherwise identical bonds denominated in different
currencies are perfect substitutes and equation (6) reduces to the familiar uncovered interest
parity relationship. Clearly, the model encompasses
the monetary models and the portfolio-
balance model, which will be delt with in section IV.4, as special cases. Real income is assumed
to be demand-determined
in the short-run. Goods-market pressure which eventually results in
price-adjustment is measured by the gap between contemporaneous and long-run equilibrium
income (equation (8)).
Natividad and Stone analyze the effects of discretionary monetary policy (increase
discretionary intervention (increase
in rFY) and of an exogenous change
in
in the foreign
~Fr,
interest rate. They note that in case of perfect substitutability between bonds denominated in different currencies and in the absence of sterilization (w
- 0) changes in monetary and exchange
rate policy by the domestic monetary authorities have identical implications (see also footnote 4).
Furthermore, they conclude that fully sterilized intervention has no effect in either the short or
long run. This result was discussed at the end of section IV.3.1. Natividad and Stone derive that
the lower the degree of endogenous sterilization x. , the larger the initial jump in the exchange
rate after a discretionary intervention (positive value of rF.). ln other words, they find that the
effect of an ad hoc policy measure by the central bank depends on the exact shape of its own
mechanical reaction pattem. This result, which is not too straightforward, is mainly due to the
implausible intervention reaction function. It is not clear why the central bank would feel the
need to intervene discretionarily given that the intervention reaction function already accounts
for "endogenous" intervention in response to observed movements in the exchange rate. More
interestingly, Natividad and Stone find that the et~ect of the degree of sterilization on the degree
(Partly) sterilized intervention in Natividad and Stone's (1990) sticky-price model
r - rFr - ~ I(s - p ~ p ' ) - (s - P ' P ~ ))
intervention reaction function
C- CFY - w~ r t wZ (l - Í)
domestic credit policy
0 5 w~ 5
1
m -h~c thzr
money supply
m~-p-~Y-~lr
money demand
s` -s
rational expectations
i- i' t s` -(I~a) (nÍa - n.ja`~
d -S(s -p fp') ~yy-ar
"uncovered interest parity"
p -rz(d -y)
price adjustment
aggregate demand
15
of overshooting after a discretionary intervention is ambiguous. A lower degree of sterilization
increases both the initial jump in the exchange rate ànd its new long-run equilibrium value.
IV.4 The portfolio balance model
IV.4,1
The standard portfolio balance model and intervenlion
The asset market models for exchange
rate detetmination
analyzed hitherto assume
domestic and foreign assets to be perfect substitutes. Portfolio balance models explicitly leave
open the possibility that risk-averse investors believe assets denominated in different currencies
to have different risk characteristics, and, hence, that they want to be compensated for the higher
perceived risk of holding foreign assets. When investors demand a nonzero risk premium on the
domestic asset (RP`~ then a wedge is driven between the expected rates of return on domestic
and foreign bonds. The uncovered interest pariry relationship no longer holds:
(0)
uncovered interest parity adjusted for a risk premium
i- i' t s` . RPD
In a world in which bonds denominated in different currencies are imperfect substitutes, the
requirement of continuous money- ànd bond market equilibrium jointly determines the exchange
rate and interest rates.
An elementary
small country portfolio model of the exchange
rate
is presented
in
Branson, Halttunen and Masson (1977). There are three assets: domestic and foreign bonds and
domestic money. In accordance with the short-run nature of the model accumulation of foreign
bonds through
current account
surpluses is ruled out." Furthermore, there
is no interaction
between the financial markets and the goods market. Therefore, the latter is not specified in the
model. The demand for money (.4~j, domestic bonds (B) and foreign bonds (SF, expressed in
domestic currency) are assumed to depend upon wealth, the own rate of return and the cross
rates whereby the rate of retum on (narrow) money is set to zero and expectations are static. To
bring down the value of the domestic currency, the domestic central bank can conduct an official
purchase of foreign bonds in exchange for domestic money. In the framework of a portfolio
balance model, this leads to an excess demand for foreign bonds and an excess supply of money.
For given levels of the exchange rate the domestic interest rate has to decline in order for the
investors to willingly hold the available stock of money. The open-market operation leaves the
market for domestic bonds unaffected
while the proportion
of wealth
held
in the form of
domestic bonds is unchanged. In the new equilibrium alI three markets have to be in equilibrium
again. The interest rate is lower so as to clear the excess supply of money (substitution effect).
The value of foreign curtency is higher in order to bid up the proportion of wealth held in the
form of foreign bonds (wealth effect).
"Stock-flow interaction in portfolio models of the exchange rate is discussed in section IV.S.
16
The effect of an unsterilized intervention in the framework of the portfolio balance
model is rather straightforward, as is the case in other models for exchange rate determination. It
neither depends crucially on the small-country assumption or the assumption of static expectations nor on the degree
changes
of substitutability between domestic and foreign
when one considers the effect
of a sterilized
intervention
bonds. The picture
in the framework
of the
portfolio balance model. Imagine the domestic central bank wants to bring down the value of the
domestic currency without altering the domestic money supply. To do so, it can buy foreign
currency-denominated
bonds from
the
private
sector
and
at
the
same
time sell
domestic
currency-denominated bonds to the private sector, leaving total private sector wealth unchanged.
How exactly are investors going to react given that their portfolio shares are not infinitely
sensitive to changes in expected rates of retum on domestic and foreign bonds?
With the economy initially in equilibrium, the swap of domestic bonds for foreign bonds
in the portfolio of the private sector leads to an excess supply of domestic bonds and an excess
demand
for foreign bonds. The money market remains in equilibrium with both the money
supply and total private sector wealth unchanged. In the new equilibrium the domestic interest
rate is higher clearing the excess supply of domestic bonds. The value of foreign currency is
higher bidding up the proportion of wealth held in the form of foreign bonds therewith wiping
out the initial excess demand for foreign bonds. Thus, the sterilized purchase of foreign bonds in
exchange for domestic bonds leads to a depreciation of the domestic currency.
Many authors find it hard to clarify the underlying mechanism by which a pure intervention alters the exchange rate; especially when the assumption of static expectations and the small
country assumption are dropped. The excess demand (supply) for (of) foreign ( domestic) bonds
causes the price of these bonds to rise ( decline) and, hence, causes the foreign
(domestic)
interest rate to decline ( rise). Obviously, although these interest rate changes make domestic
bonds relatively more attractive than foreign ones, they are mere mechanical reactions to changes in the supply-conditions on the markets for both rypes of bonds. According to Henderson
and Sampson ( 1983), given imperfect substitutability, investors require an additional inducement
to switch their foreign bonds for domestic bonds. The authors have in mind the uncovered
interest parity relation adjusted for a risk premium (equation ( 0). In addition to the mechanical
price changes the expected rate of retum on domestic bonds ( r) has to rise relative to that on
foreign bonds (i' }(ES,,, - s,l) to enhance the attractiveness of domestic bonds. Thus, with the
expected exchange rate (E~s,,,) assumed coristant, the value of domestic currency has to decline
initially (rise in s,) in order to orchestrate an expected appreciation of the domestic currency.'1
"In case of perfect substitutability the latter exchange rate change does not occur. Furthermore, the demand for bonds denominated in either currency are perfectly elastic. Hence, after
the swap of domestic for foreign bonds only an incipient rise (decline) of the domestic (foreign)
interest rate is sufficient to restore equilibrium in all three financial markets.
17
IV.-1.2
Unsterilized intervention in the portfolio model by Moreno and Yin (1992)
Until now, the effectiveness of intervention working through the monetary channel is
undisputed. Moreno and Yin (1992) come up with some new theoretical insights. They draw on
the experience of Taiwan in the 1980s which tried to limit fluc[uations in the New Taiwan (NT)
dollarN.S.
dollar-exchange
rate. The authors develop a small country portfolio
model with
flexible prices. There are three assets: domestic money (m), domestic bonds ( b) and foreign
bonds (n. The real demand for each of these assets is a function of the nominal return on domestic bonds, the expected rate of depreciation of the domestic currency (ES,., - s,), real wealth
(w) and real income (y). The small country assumption implies a fixed foreign interest rate. In
the model it is normalized at zero. Central to the analysis by Moreno and Yin are the reduced
fonn responses of the exchange rate and the price level to shocks to the exogenous variables.
(1)
ds, - s, 0(E~S,., - s,) t sT t1 m,' ~ s~ ~ f,'
s, ~ 0, s,~ ~ 0, s~ ~ 0
(2)
OP, - P~ 0(E~,.~ - s,) t p~, 0 m,' t p~ ~.i~t
p, ' 0, Pm ' 0, P~ ~ 0
whereby
(3)
~s~P - s p(ES ~ - s)
change in expectations of private sector
(4)
~ si` B--~ 0 s,P -- i; s, 4(E~S„~ - s,)
intervention rule, with 0 5 S ~ 1
Moreno and Yin trace out the effect of a shock which causes an expected appreciation of domestic currency. The expected rise in the value of domestic currency (0 (E~ i- s) ~ p) increases
the demand for assets denominated in domestic currency. As a result, the domestic currency
appreciates (equation ( I)). The authors analyze the extent to which the central bank is able to
neutralize this appreciation by conducting unsterilized purchases of foreign bonds according to
the rule in equation (4). The intervention operation leads to an increase in the domestic money
supply and a matching reduction in the supply of foreign assets held by domestic residents. As a
result, the initial appreciation of domestic currency is (partly) reversed. The authors arrive at the
familiar insight that the domestic central bank can limit exchange rate changes to any degree
desired (by chosing a high value of ~) but only at the cost of larger changes in its holdings of
net foreign assets and, hence, in the domestic money supply. Moreno and Yin point out that the
story not neccesarily ends here. The exchange rate change to which the central bank reacted was
initiated by a shock to private investors' exchange rate expectations.
In the terminology of the
authors the intervention outcome is credible when investors believe that, after the monetary
intervention in reaction to the expectations' induced change in the exchange rate, there will be
no further change in the exchange rate. By contrast, the outcome of intervention is not credible
when investors "believe that the exchange rate must ultimately adjust to some target exchange
rate s', regardless of the short-run attempts of policymakers to prevent such adjustment" (p. 20).
As
before,
attempts
by
the central
bank to prevent (part of)
the current exchange
rate's
adjustment to the target level will in first instance reverse (part of) the initíal appreciation of the
18
domestic currency. However, the resulting level of the exchange rate lacks credibility and the
expectation of an appreciation of domestic currency persists as long as s~ s'. This leads to
renewed excess demand for assets denominated in domestic currency. It depends on the intensity
of unsterilized intervention how long it will take exactly but, as the reserves of a central bank
are finite, the exchange rate will ultimately become equal to the target rate investors have in
mind. Of course,
in the absence of intervention
(~ - p) the full impact of the shock to
expectations is felt in the first period. A higher value of ~ leads to a distribution of the impact
of the one-time shock over time at the cost of higher cumulative intervention.
At first sight, the persistence in the appreciation-expectation
does not seem very plausi-
ble. The unsterilized interventions lead to an increase in the domestic money supply and this will
eventually lead to a rise in domestic inflation. For an open economy which has a competitive
advantage over other countries, a rise in domestic prices is a close substitute for an appreciation
of the domestic currency; both detract from the competitiveness of domestic industries and,
hence, will go a long way in removing the cause of the initial expected appreciation of domestic
currency. Moreno and Yin come up with an interesting explanation which also sheds a new light
on the exposition in section 11. In the short-run, unsterilized intervention can prevent equilibrium
exchange rate adjustment. This may result in a sequence o( growing trade surpluses for the
domestic country. At the same time, investors may revise their estimates of s,' upward when
confronted with news indicating no reduction
in the trade surplus or complaints by trading
partners." The unsterilized purchases of foreign bonds in exchange for domestic money both
have an inflationary effect on the domestic economy whereas the recurrent
updating of the
expected appreciation has a deflationary effect." Thus, the increase in the money supply resulting from intervention will be associated with a less than proportional increase in inflation. The
innovations in the appreciation-expectation
hamper the adjustment proces ~vhich underlies the
working of the monetary channel of intervention. Moreno and Yin conclude from the analysis
that the persistent and accelerating appreciation of the NT dollar in the 1980s may have been
related to government efforts to limit such appreciation which lacked credibility.
IV.-0.3
Sterilized intervention in Blundell-Wignall and Masson's (19851 mode!
"Note that a subscript t is attached to this expression for the target exchange rate. This
indicates that the curtent analysis is concemed with a sequence of negative shocks to exchange
rate expectations rather than a one-time expected appreciation of the domestic currency. The
updating of the target rate by investors in this case implies that E~s,., ~ E,.,s„1 ~ E„ls,.,, etc.
Id
Due to the intervention: ~ f,` ~ 0, from equation (2) with p~ ~ 0 it follows that ~ p, ~ 0
Due to the intervention: A m,s ~ 0, from equation (2) with p~, ~ 0 it follows that 4 p, ~ 0
Revision of expectations: 0(E~S„~ - s,) ~ 0, from (2) with p, ~ 0 it follows that 0 p, j 0
19
In fact; there is only a minor difference between the sticky-price monetary model and the
portfolio model of exchange rate determination. Blundell-Wignall and Masson ( 1985) extend
Dornbusch's ( 1976) model to include assets denominated in domestic and foreign currencies that
are not perfect substitutes. Equation ( 4) reflects that, in order to be induced to hold more of
foreign assets, domestic investors require a higher expected retum on them. The risk premium on
foreign assets is assumed to depend on the private stock of net claims on foreigners (nfa - njal~.
Blundell-Wignall and Masson assume that the central bank systematically tries to resist movements in the real exchange rate away from a constant long-run equilibrium value which is, moreover, publicly known. When the current real exchange rate is above its long-run value SR the
central bank will enter the foreign exchange market to buy domestic currency ( ~i , 0). Furthermore, the central bank tries to prevent the stock of reserves from deviating too far from some
target Ievel ~ ( equation ( 1)). A country which experiences a current account surplus accumulates
foreign currency-denominated assets. StoclJflow-interaction is not incorporated in the model, i.e.
the current account and the net foreign asset-position are exogenous. According to equation (7),
the current real exchange rate may differ from its equilibrium level either because real interest
rates differ at home and abroad or because private net claims on foreigners are nonzero. Obviously, sterilized purchases of foreign bonds affects the volume of private net claims on foreigners and, hence, have a bearing on the current real exchange rate through the portfolio channel
described at the end of section 1V.4.1. Blundell-Wignall and Masson establish that the model is
stable whatever the value of the intervention parameter ~. It follows that the rule for sterilized
intervention in equation ( I) allows the monetary authorities "to guide the exchange rate toward
its long-run equilibrium value without inducing short-run fluctuations in that rate" ( p. 140).
From equation ( 3) it appears that private exchange market participants either do not anticipate the intervention behaviour or, if they do, consider it has no effect. To correct for this
feature, Blundell-Wignall and Masson modify the model so as to let expectations correctly take
account of intervention. With fully rational expectations regarding the exchange rate, equation
Sterilized intercention in Masson and Blundell-Vb'ignall's (1985) sticky-price model
SR - S -P t p.
r - ~I
(SR
-
SR)
definition of the real exchange rate
t ~. (r - r)
m - P t ~Y - ~ i
SR
-
e (SR
- SR)
intervention reaction function
money market equilibrium
expected real depreciation
i-i' ts`-(1~P)(nfa-nÍa``i
d - SsR t yy -Q (i -p`)
"uncovered interest pariry"
p -n(d -y) tp`
price adjustment
SR - SR ' ~(i ' - p ' `) - (i - p `)]IA - (nja - r)l6(3
aggregate demand
SOIUt10R
20
(3) reduces to s~- s and equation (4) can be rewTitten as s- i- r~ .(I~p) (nfa - r), 7~e
authors show that the resulting model has the saddle-point property and, provided that it is
sufficiently strong, sterilized intervention does not lead to cyclical fluctuations in the exchange
rate. The mechanics of sterilized intervention in the amended Dornbusch model are not made
clear entirely. Shocks cause short-run misalignments of the real exchange rate but leave the
long-run equilibrium value unafiected. The famous Dornbusch-overshooting refers to nominal
exchange rates overshooting their long-run value in response to an increase in the domestic
money supply. With prices sticky in the short-run, the real exchange rate also overshoots.
Sterilized intervention, supposedly working through the channel due to Henderson and Sampson
(1983) described above is said to "speed up adjustment to past shocks" and to "help lessen
overshooting" [Blundell-Wignall and Masson (1985, p. 142)].
IV.S Stock-flow interaction in portfolio models of excóange rate determination
The portfolio model discussed
in section IV.4 are essentially short-run asset market
models. There is no interaction between the asset markets and the goods market. Furthermore, in
accordance with the short-run nature of the model accumulation of foreign assets through current
account surpluses is ruled out. However, the instantaneous restoration of equilibrium on the
financial markets after a shock to the system involves an adjustment of the exchange rate. This
change in the value of foreign exchange alters the ratio of domestic to foreign goods prices
expressed in a common curtency. The trade balance improves or deteriorates and, hence, the
current account of the balance of pa}ments will show a surplus or a deficit. In tum, this will
cause an accumulation or a decumulation of foreign assets in the hands of the domestic private
sector. It follows that, beyond the short-run, the course of the exchange rate is determined by
current account and capital account flows of foreign exchange which induce an adjustment of the
stocks of foreign and domestic assets in the hands of the private sector. The exchange rate
moves until the current account and the capital account are again individually in equilibrium.
Dornbusch and Fischer (1980) is a seminal article on long-term adjustment processes and
the concomitant interaction between flows of foreign exchange and the dynamics of asset stocks.
Current- and capital account dynamics and the adjustment path of the exchange rate after an
unsterilized and a sterilized intervention can be studied in the framework of Branson (1983) and
Hallwood and MacDonald (1986, Chapter 7). [n section IV.4 it was established that both the
monetary and the pure intervention lead to an instantaneous depreciation of the domestic currency. As a consequence, a current account surplus emerges and the domestic economy accumulates additional foreign assets. The current account surplus puts upward pressure on the value of
domestic currency. Therefore, in the course of the adjustment to the new steady state the domestic currency appreciates and the rate of accumulation of net foreign assets diminishes gradually
to zero. The new long-run equilibrium consequent upon the initial monetary or pure intervention
must be one in which the nominal exchange rate has risen.
21
V ALTERNATIVE APPROACHES TO THE STUDY OF FOREX INTERVENTION
V.I Introduction
The theoretical investigations discussed in the previous sections all assumed that some
structural model of exchange rate determination provides a valid framework for the analysis of
the
effectiveness
of foreign
exchange
market
intervention.
After
surveying
the
empirical
evidence on exchange rate models, MacDonald and Taylor (1992, p. 24) conclude that "...the
asset approach models have perfortned well for some time periods, such as the interwar period,
and, to some extent, for the first part of the recent floating experience (that is, 1973-1978); but
they have provided largely inadequate explanations for the behavior of the major exchange rates
during the latter part of the float". Discontent with the performance of traditional structural
exchange
rate models in explaining the actual
behaviour of exchange
rates has led many
economists to adopt new research strategies in exploring the field of exchange rate economics.
V.2 De Grauwe's (1989) near-rationality model
De Grauwe (1989) emphasizes the crucial role of uncertainty among foreign exchange
market
participants
regarding
the
future
course of currency
movements.
In his
integrated
approach, noise trading behaviour seems to follow directly and in a consistent way from trading
strategies based on fundamental economic analysis which are assumed to prevail in the structural
models of exchange rate determination examined in the previous sections. De Grauwe argues
that economic models providing a reliable guide for forecasting the future exchange rate are
lacking. Within a mean variance framework he shows that it is not necessarily profitable to use
all available fundamental economic information. For instance, an individual exchange market
participant may observe a gap between the actual level of the exchange rate and the perceived
fundamental equilibrium value. In a highly uncertain economic environment, the expected gain
from taking a forward position aimed at exploiting this gap may not outweigh the risk involved.
Therefore, it can be rational for exchange market participants to implement technical rather than
fundamental analysis. De Grauwe presents a rule for the formation of exchange rate expectations
that consists of a backward-looking and a forward-looking component, the first term and the
second term on the right-hand side of (1), respectively:
(1)
EPS,.i - k ( ~ c,cS,-,) i (1 -k) (S ' -S~)
~-o
The parameter k which represents the weight given to each of the expectation rules is assumed
to be endogenous. Exchange market participants choose to let the backward-looking component
dominate their expectations formation (high value of k) when the actual exchange rate is not too
22
far away from the equilibrium rate they have in mind (S'). This implies that they resort to
technical analysis. A higher weight is assigned to the forward-looking component (lower value
of k) only in case it is obvious that the actual exchange rate is at an unsustainable level. Foreign
exchange trading based on fundamental analysis becomes less risky leading to a larger expected
utility in a mean variance framework. By assuming that exchange market participants revert to
backward-looking technical analysis instead of forward-looking fundamental analysis in case of
extreme uncertainry, De Grauwe's near rationality model can account for some stylized facts of
(real)
exchange
rates which are left unexplained
by
perfect
foresight rational
expectations
models, i.e. real exchange rates wandering away from fundamentals for long periods of time
before being pushed back towards equilibrium and the relative sluggishness of exchange rate
movements as compared to stock and commodiry price movements."
While the current global exchange rate system offers no credible anchor for exchange
rate expectations,
De Grauwe (1989) sees no role for ad hoc (un)sterilized
intervention
to
remedy persistent deviations of exchange rates from their perceived equilibrium values. In his
view, even changes in fundamentals brought about by monetary or fiscal policy will not help.
Due to the extreme uncertainty as to the true model, market participants do not know how to
interpret these changes. De Grauwe argues that only a credible exchange rate commitment by the
monetary authorities, i.e. a commitment embodying a clear and stabilizing guide for exchange
market participants' expectations, can facilitate an efficient functioning of the foreign exchange
market comparable to stock and commodiry markets. This would require systematic unsterilized
interventions rather than ad hoc (un)sterilized central bank operations in foreign exchange with
which the analysis in the previous sections was concemed. In sum, De Grauwe takes account of
non-fundamentalist
behaviour or, to be more precise,
chartist
behaviour by (some) foreign
exchange market participants. This does not open up a distinct channel through which foreign
exchange intervention can affect the course of currency movements.
V.3 Chartist channel and noise trading signaling channel of intervention
Fukao (1985) argues that the scope for intervention changes quite dramatically when one
is willing to drop the assumption that the foreign exchange market is efficient in the sense that
"In an attempt to ascertain the extent and manner by which chartism is used Allen and
Taylor (1989) conducted a questionnaire survey of chief foreign exchange dealers in the London
market. They found, inrer alio, that "at the shortest horizons, intraday to one week, approximately 90a~o of respondents use some chartist input in forming their exchange rate expectations, with
60"~a judging charts to be at least as important as fundamentals. At longer forecast horizons, of
one to three months or six months to one year, the weight gíven to fundamentals increases..." (p.
50). Furthermore, ín a recent study De Grauwe and Decupere ( 1992) cannot reject the hypothesis
that psychological barriers exist in the yen-U.S. dollar market. For the DM-U.S. dollar market
the results are mixed.
23
market participants
use some structural
model as a yardstick
when taking positions on the
foreign exchange market. Fukao presents the following table which is both simple and insightful.
Table 2. Effectiveness of Intervention and the state of the foreign exchange market
SubstitutabiliR of bonds denominated in different currencies
Perfect
Im erfect
Efficient Market
Intervention is ineffective
Intervention is effective
Inefficient Market
Intervention is effective
Intervention is effective
Source: Fukao (1985, p.26)
Fukao (1985) does not clarify how the inefficiency of the foreign exchange market opens
up a distinct channel through which foreign exchange intervention can affect the exchange rate.
Hung (1991a,b) contends that the presence of non-fundamentalist
trading behaviour is more or less predictable
intervention can be effective.
Hung's approach
market participants whose
constitutes a channel through
which sterilized
is slightly different from that of De Grauwe
(1989). The noise trading behaviour by exchange market participants is not derived formally.
Earlier work on noise trading is just put to the right use. As De Grauwe, Hung arrives at the
important insight that fundamentalists can (at times) tum into noise traders due to the extreme
uncertainty regarding the future path of the exchange rate. She distinguishes two categories of
non-fundamentalist or noise traders. Chartists are assumed to '...rely on analysis of past price
pattems to predict the future direction of exchange rate movements' (p. 12). Market participants
who base trading on their prediction of the market's reaction to news and rumors are called nonchartist norse traders. The analysis is conducted in the framework of a partial equilibrium flow
market approach to exchange rate determination. If a large enough group of noise traders use the
same forecasting technique or share the same belief with respect to the future course of the exchange rate, expectations may tum out to be self-fulfilling. The central bank may want to counter the resulting curtency movements. Hung argues that the noise trading behaviour displayed by
market participants in itself offers an opportunity for the central bank to do this successfully.
A necessary but of course not sufficient condition for intervention to be effective is that
the central bank can imagine itself in the position of the two groups of noise traders distinguished. Current intervention volumes are insutiicient to counter a strong underlying trend in the
exchange rate. Therefore, the decision when and how to intervene has to be made conditional on
information about the general market sentiment and the (strength of the) buy or sell signals
adherents of technical anal}~sis derive from their charts. Hung (1991a, p. 20) presents an ideal
24
picture in which chartists "... help amplify and prolong the effect of intervention" which may be
transitory by itself The central bank should select a time period during which the market is
sufficiently thin. Concealed intervention carried out through brokers may cause just enough
up~vard or downward pressure on the currency under consideration which, if incorporated in the
chartists' trendline analysis induces them to reinforce the movement of the exchange rate in the
direction favoured by the central bank. This is what Hung calls the chartist channel of inlervention. When the interventions indeed remain anonymous, the central bank has nothing to loose in
terms of reputation. The latter condition may account for the fact that detailed intervention data
are not made publicly available. On the other hand, Hung (19916, p.7) admits that "...chart
analysis has a large subjective element, and there are probably as many methods of combining
and interpreting the various techniques as there are chartists themselves". A second channel is
the noise rrading signaling channel. For this channel, the opportuniry to influence the course of
the exchange rate does not lie in the thinness of the market or the fact that traders only stare at
their charts.
For the noise trading signaling channel
to be effective,
noise traders must be
"...already looking for any sign or excuse to reverse their position" [Hung (1991b, p.
12)].
Highly visible interventions, conducted via the interbank market could give just such a signal.
VI CONCLUSIONS
It has been the aim of this study to provide a comprehensive picture of theories on the
scope for foreign exchange market intervention and to find out whether and how intervention is
able to influence the course of the exchange rate. In section II a definition of exchange market
intervention is given. Unsterilized interventions conducted by the domestic central bank cause a
change in the domestic monetary base. By contrast, sterilized interventions lack a money market
effect. They do lead to a change in the currency composition of private investors' portfolios.
The effectiveness of monetary intervention in the various models studied in section [II
and IV is undisputed. Even elementary purchasing power parity predicts that an unsterilized
intervention will in the long run, ceteris paribus, influence the level of the exchange rate in a
one to one relationship with a purchase of foreign curtency by the domestic central bank leading
to a rise in the value of foreign exchange in terms of the domestic currency. In the theoretical
models investigated in section lll and IV, sterilized intervention can only be effective in case
risk-averse investors believe assets denominated
in different currencies to have different risk
characteristics, and, hence, that they want to be compensated for the higher perceived risk of
holding either domestic or foreign assets. In that case, a swap of domestic bonds for foreign
bonds in the portfolio of the private sector brought about by a sterilized purchase of foreign
currency by the domestic central bank can lead to an appreciation of foreign currency. The
practical relevence of this channel of influence is an empirical issue.
Attention is also paid to the popular insight that sterilized intervention can alter the
2s
current exchange rate through what is called the expectations or signalling channel of intervention by providing private exchange market participants with new information or a signal about the
future course of monetary
policy.
The paper develops the argument
that for the signalling
hypothesis to be valid central banks have to back up interventions with subsequent changes in
monetary policy. In other words, current sterilized interventions pre-determine the path of future
money
grownh
intervention's
and
money
hence
interfere
market effect
with monetary policy.
is limited to the
The
short
neutralization
of the
initial
run. This no longer meets
the
definition of sterilized intervention presented earlier in the paper.
From section V it appears that the scope for intervention changes quite dramatically
when one is willing to drop the assumption that the foreign exchange market is efficient in the
sense that market participants use some structural model as a yardstick when taking positions on
the foreign exchange market.
However, the literature on how the inefficiency of the foreign
exchange market opens up distinct channels through which intervention can affect the exchange
rate is still in its infancies.
26
REFERENCES
Allen, H., and M.P. Taylor, "Charts, Noise and Fundamentals in the Foreign Exchange Market",
Economic Journal, 1990, 100 (supplement): 49-59.
Almekinders, G.J. and S.C.W. Eijffinger, "Empirical Evidence on Foreign Exchange Market
Intervention: Where Do We Stand?", Weltwirtschaftliches Archiv, 1991, 127: 645-677.
Alogoskoufis, G. "On Inflation, Employment and the Optimal Exchange Rate Regime", in F. van
der Ploeg (ed.), Handbook of International Macroeconomics, Oxford 1993, forthcoming.
Belongia, M.T., "Foreign Exchange Interventíon By the United States: A Review and
Assessment of 1985-1989", Review, Fed. of St. Louis, MaylJune 1992, 74: 32-51.
Black, S.W., "The Effect of Alternative
Intervention Policies on the Variabilit}
of Exchange
Rates: The Harrod Effect", in J.S. Bhandari (ed.), Exchange Rate Management Under
Uncertainty, MIT Press, Cambridge MA, 1985.
Blundell-Wignall, A. and P.R. Masson, "Exchange Rate Dynamics and Intervention Rules", Staff
Papers, International Monetary Fund, 1985, 31: 132-159.
Branson, W.: "Macroeconomic deterrninants of real exchange risk", in: R.1. Herring (ed.),
Managing foreign exchange risk, Cambridge University Press, 1983, p. 33-74.
Branson, W.H., H. Halttunen and P. Masson: "Exchange Rates in the Short Run-The DollarDeutschemark Rate", European Economic Review, 1977, 10: 303-324.
Djajic, S. and S. Bazzoni, "Dynamics of Monetary Intervention under Managed Floating",
.
European Economic Review, 1992, 36: 1399-141 I.
Dornbusch,
R., "Expectations and Exchange Rate Dynamics", Journol of Political Economy,
1976, 84: I 161- I 176.
Dunnen, E. den and S. de Wilde, Instruments of money market and foreign exchange market
policy in the Netherlands, NIBE, 1992.
Edison, H., "The Effectiveness of Centra! Bank In(ervention: A Survey of the Post-1982
Literature", manuscript, Board of Govemors of the Federal Reserve System, June 1992.
Fleming, J.: "Domestic Financial Policies under Fixed and under Floating Exchange Rates",
International Monetary Fund Staff Papers, 1962, 2: 347-354.
Fukao, M., "The Effectiveness of Coordinated Intervention", Policy StuáV Series, No. 3, April
1985, Japan Center for International Finance.
Giovannetti, G., "A Survey of Recent Empirical Tests of the Purchasing Power Parity
Hypothesis, Banca Nazionale Del Lavoro Quarterly Review, 1992, No. 180, 81-101.
Grauwe, P. de, Internationa! Money, Post-War Trends and Theories, Oxford, 1989.
Grauwe, P. de, and D. Decupere,
"Psychological Barriers In The Foreign Exchange Market",
Centre for Economic Policy Research, Discussion Paper, No. 621, January 1992.
Hallwood, P. and R. MacDonald: International .Lfoney,
London, 1986.
Theory, Evidence and Institutions,
27
Henderson, D.W. and S. Sampson: "lntervention in Foreign Exchange Markets: A Summary of
Ten Staff Studies", Federal Reserve Bulletin, November 1983, 69: 830-836.
Humpage, O. F., "Exchange
Market Intervention: The Channels of Influence", Economic
Review, Federal Reserve Bank of Cleveland, 1986, Vol. 22, No. 3, 2-13.
Hung, J.H.,(1991a) "Noise Trading and the Effectiveness oJSterilized Foreign Exchange
Intervention", Fed of New York, Research Paper No. 9111 (Mazch 1991).
Hung, 1.H.,(1991b) "The Effectiveness OjSterilized U.S Foreign Exchange lntervention; An
Empirical Stuáy Based On The Noise Trading Approach", Federal Reserve Bank of New
York, Research Paper No. 9118 (May 1991).
Kaminsky, G.L. and K.K. Lewis: "Does Foreign Exchange Intervention Signa! Future Monetary
Policy", manuscript, Board of Govemors, Federal Reserve System, December 1992.
Klein, M.W., and E.S. Rosengren, "Foreign Exchange Intervention as a Signal of Monetary Policy", New Englond Economic Review, Fed of Boston, May~June 1991: 39-50.
MacDonald, R., Floating Exchange Rates-Theories and Evidence, London, 1988.
MacDonald, R. and M.P. Taylor, "Exchange Rate Economics: A Survey", Stojf Papers, International Monetary Fund, Mamh 1992, 39: I-57.
Marston, R.C., "Stabilization Policy in Open Economies", in: Jones R.W. and P.B. Kenen (ed),
Nandbook of lnternational Economics, vol. 11, North-Holland, Amsterdam.
Moreno, R. and N. Yin, "Exchange Rate Policy and Shocks to Asset Markets: The Case of
Taiwan in the 1980s", Economic Review, Fed. of San Francisco, 1992: 14-34.
Mundell, R. A., "Capital Mobility and Stabilization Policy under Fixed and Flexible
Exchange Rates, Canadian Journal oj Economics and Political Science,
1963,
29: 475-485.
Mussa, M.: "The Role of Official Intervention", Group of Thirty Occasional Papers, No. 30,
New York, 1981.
Natividad, F., and Stone, J.: "A General Equilibrium Model of Exchange Market Intervention
with Variable Sterilization", Journal oJlnternational Economics, 1990, 29: 133-145.
Neumann, M.J.M., "Intervention in the MarklDollar Market: the Authorities' Reaction Function",
Journal oJlnternationa! Money and Finance, 1984, 3: 223-239.
Okun, A.M., Prices and Quantities: a macroeconomic analysis, Oxford, Blackwell, 1981.
Pilbeam, K., Exchange Rate Management: Theory and Evidence, Basingstoke, MacMillan, 1991.
Visser, H., "Exchange Rate Theories", De Economist, March 1989, 137: 16-46.
Weber, W.E., "Do Sterilized Interventions Affect Exchange Rates?", Quarterly Review,
Federal Reserve Bank of Minneapolis, Summer 1986, p. 14-23.
Discussion Paper Series, CentER, T[lburg Unlversity, The Netherlands:
(For previous papers please consult previous discussion papers.)
No.
Author(s)
Title
9206
W. H~rdle
Applied Nonparame[ric Models
9207
M. Verbeek and
Th. Nijman
Incomplete Panels and Selection Bias: A Survey
9208
W. H~rdle and
A.B. Tsybakov
How Sensitive Are Average Derivatives?
9209
S. Albark and
P.B. Overgaard
Upstream Pricing and Ach"ertising Signal Downstream
Demand
9210
M. Cripps and
J. Thomas
Reputation and Commitment in Two-Person Repeated
Games
9211
S. Albak
Endogenous Timing in a Game with Incomplete Information
9212
T.J.A. Storcken and
P.H.M. Ruys
Extensions of Choice Behaviour
9213
R.M.W.J. Beetsma and
F. van der Ploeg
Exchange Rate Bands and Optimal Monetary Aocommodation
under a Dirty Float
9214
A. van Soest
Disaete Choice Models of Family Labour Supply
9215" W. Giith and
K. Ritzberger
On Durable Goods Monopolies and the (Anti-) CoaseConjecture
9216
A. Simonovits
Indexation of Pensions in Hungary: A Simple Cohort Model
9217
J.-L. Ferreira,
I. Gilboa and
M. Maschler
Credible Equilibria in Games with Utilities Changing
During the Play
9218
P. Borm, H. Keiding,
R. Mclean, S. Oortwijn
and S. Tijs
The Compromise Value for NTU-Games
9219
J.L. Horowitz and
W. Hárdle
Testing a Parametric Model against a Semiparametric
Alternative
9220
A.L. Bovenberg
Investment-Promoting Policies in Open Economies: The
Importance
of
Intergenerational
and
International
Distributional Effects
9221
S. Smulders and
Th. van de Klundert
MonopoGstic Competition, Product Variety and Growth:
Chamberlin vs. Schumpeter
9222
H. Bester and
E. Petrakis
Price Competition and Advertising in Oligopoly
No.
Author(s)
Title
9223
A. van den Nouweland,
M. Maschler and
S. Tijs
Monotonic Games are Spanning Network Games
9224
H. Suehiro
A "Mistaken Theories" Refinement
9225
H. Suehiro
Robust Selection of Equilibria
9226
D. Friedman
Economically Applicable Evolutionary Games
9227
E. Bomhoff
Four Econometric Fashions and
Alternative - A Simulation Study
9228
P. Borm, G.-J. Otten
and H. Peters
Core Implementation in Mod~ed Strong and Coalition Proof
Nash Equilibria
9229
H.G. Bloemen and
A. Kapteyn
The Joint Estimation of a Non-Linear Labour Supply Function
and a Wage Equation UsingSimulated Response Probabilities
9230
R. Beetsma and
F. van der Ploeg
Does Inequality Cause Inflation? - Tfie Political Economy of
Inflation, Taxation and Government Debt
9231
G. Almekinders and
S. Eijffinger
Daily Bundesbank and Federal Reserve Interventions
- Do they Affect the Level and Unexpected Volatility of the
DM~S-Rate?
9232
F. Vella and
M. Verbeek
Estimating the Impact of Endogenous Union Choice on
Wages Using Panel Data
9233
P. de Bijl and
S. Goyal
Technological Change in Markets with Network Externalities
9234
J. Angrist and
G. Imbens
Average Causal Response with Variable Treatment Intensity
9235
L. Meijdam,
M. van de Ven
and H. Verbon
Strategic Decision Making and the Dynamics of Government
Debt
9236
H. Houba and
A. de Zeeuw
Strategic Bargaining for the Control of a Dynamic System in
State-Space Form
9237
A. Cameron and
P. Trivedi
Tests of Independence in Parametric Models: With
Applications and Illustrations
9238
J.-S. Pischke
Individual Income, Incomplete Information, and Aggregate
Consumption
9239
H. Bloemen
A Model of Labour Supply with Job Offer Restrictions
9240
F. Drost and
Th. Nijman
Temporal Aggregation of GARCH Processes
the
Kalman
Filter
No.
Author(s)
Title
9241
R. Gilles, P. Ruys
and J. Shou
P. Kort
Coalition Formation in Large Network Economies
9243
A.L. Bovenberg and
F. van der Ploeg
Env'uonmental Policy, Public Finance and the Labour Market
in a Second-Best World
9244
W.G. Gale and
J.K. Schotz
IRAs and Household Saving
9245
A. Bera and P. Ng
Robust Tests for Heteroskedasticity and Autocorrelation
Using Score Function
9246
R.T. Baillie,
C.F. Chung and
M.A. Tieslau
The Long Memory and Variability of Inflation: A
Reappraisal of the Friedman Hypothesis
9247
M.A. Tieslau,
P. Schmidt
and R.T. Baillie
A Generalized Method of Moments Estimator for LongMemory Processes
9248
K W~rneryd
Partisanship as Information
9249
H. Huizinga
The Welfare Effects of Individual Ret'uement Accounts
9250
H.G. Bloemen
Job Search Theory, Labour Supply and Unemployment
Duration
9251
S. Eijffinger and
E. Schaling
Central Bank Independence: Searching for the Philosophers'
Stone
9252
A.L. Bovenberg and
R.A. de Mooij
Environmental Taxation and Labor-Market Distortions
9253
A. Lusardi
Permanent Income, Curcent Income and Consumption:
Evidence from Panel Data
9254
R. Beetsma
Imperfect Credibility of the Band and Risk Premia in the
European Monetary System
9301
N. Kahana and
S. Nitzan
Credibility and Duration of Political Contests and the Extent
of Rent Dissipation
9302
W. Guth and
S. Nitzan
Are Moral Objections to Free Riding Evolutionarily Stable?
9303
D. Karotkin and
S. Nitzan
Some Peculiarities of Group Decision Making in Teams
9304
A. Lusardi
Euler Equations in Micro Data: Merging Data from Two
Samples
9242
The Effects of Marketable Pollution Permits on the Firm's
Optimal Investment Policies
No.
Author(s)
Title
9305
W. Guth
A Simple Justification of Quantity Competition and the
Cournot-Oligopoly Solution
9306
B. Peleg and
S. Tijs
The Consistency Principle For Games in Strategic Form
9307
G. Imbens and
A. Lancaster
Case Control Studies with Contaminated Controls
9308
T. Ellingsen and
K. Wárneryd
Foreign Direct Investment and the Political Economy of
Protection
9309
H. Bester
Price Commitment in Search Markets
9310
T. Callan and
A. van Soest
Female Labour Supply in Farm Households: Farm and
Off-Farm Participation
9311
M.Pradhan and
A. van Soest
Formal and Informal Sector Employment in Urban Areas of
Bolivia
9312
Th. Nijman and
E.Sentana
Marginalization and Contemporaneous Ag,gregation in
Multivariate GARCH Processes
9313
K. W~rneryd
Communication, Complexity, and Evolutionary Stability
9314
O.P.Attanasio and
M. Browning
Consumption over the Life Cycle and over the Business
Cycle
9315
F. C. Drost and
B. J. M. Werker
A Note on Robinson's Test of Independence
9316
H. Hamers,
P. Borm and
S. Tijs
On Games Corresponding to Sequencing Situations
with Ready Times
9317
W. Giith
On Ultimatum Bargaining Experiments - A Personal Review
9318
M.J.G. van Eijs
On the Determination of the Control Parameters of the
Optimal Can-order Policy
'
9319
S. Hurkens
Multi-sided Pre-play Communication by Burning Money
9320
J.J.G. Lemmen and
S.C.W. Eijffinger
The Quantity Approach to Financial Integration: The
Feldstein-Horioka Criterion Revisited
9321
A.L. Bovenberg and
S. Smulders
Environmental Quality and Pollution-saving Technological
Change in a Two-sector Endogenous Growth Model
9322
K.-E. W~rneryd
The Will to Save Money: an Essay on Economic Psychology
9323
D. Talman,
Y. Yamamoto and
Z. Yang
The (2"'m" - 2)-Ray Algorithm: A New Variable Dimension
Simplicial Algorithm For Computing Economic Equilibria on
S" x Rm
No.
Author(s)
Title
9324
H. Huizinga
The Financing and Taxation of U.S. D'uect Investment
Abroad
9325
S.C.W. Eijffinger and
E. Schaling
Central Bank Independence: Theory and Evidence
9326
T.C. To
Infant Industry Protection with Learning-by-Doing
9327
J.PJ.F. Scheepens
Bankruptcy Litigation and Optimal Debt Contracts
9328
T.C. To
Tariffs, Rent Extraction and Manipulation of Competition
9329
F. de Jong, T. Nijman
and A. R6e11
A Comparison of the Cost of Trading French Shares on the
Paris Bourse and on SEAQ International
9330
H. Huizinga
The Welfare Effects of Individual Retirement Accounts
9331
H. Huizinga
Time Preference and International Tax Competition
9332
V. Feltkamp, A. Koster,
A. van den Nouweland,
P. Borm and S. Tijs
Linear Production with Transport of Products, Resources and
Technology
9333
B. Lauterbach and
U. Ben-Zion
Panic Behavior and the Performance of Circuit Breakers:
Empirical Evidence
9334
B. Melenberg and
A. van Soest
Semi-parametric Estimation of the Sample Selection Model
9335
A.L. Bovenberg and
F.van der Ploeg
Green Policies and Public Finance in a Small Open Economy
9336
E.Schaling
On the Economic Independence of the Central Bank and the
Persistence of Inflation
9337
G.-J.Otten
Characteriz8tions of a Game Theoretical Cost Allocation
Method
9338
M. Gradstein
Provision of Public Goods With Incomplete Informatibn:
Decentralization vs. Central Planning
9339
W. Guth and H. Kliemt
Competition or Co-operation
9340
T.C. To
Export Subsidies and Oligopoly with Switching Costs
9341
A. Demirgu~-Kunt and
H. Huizinga
Barriers to Portfo6o Investments in Emerging Stock Markets
9342
G.J. Almekinders
Theories on the Scope for Foreign Exchange Market
Intervention
P.O. BOX 90153, 5000 LE T~LBURG, THE NETHERLAND:
Bibliotheek K. U. Brabant
i ~ i~r~ uiu~~uii ~ i
1 7 000 O ~ ~ 63438 4